Nature of Nonbonded Se...O Interactions Characterized by 17O NMR Spectroscopy and NBO and AIM Analyses

Article abstract of DOI:10.1021/ja049690n

To investigate the nature of nonbonded Se...O interactions, three series of 2-substituted benzeneselenenyl derivatives [2-(CHO)C₆H ₄SeX (1), 2-(CH₂OH)C₆H₄SeX, (2), 2-(CH₂OPr)C₆H<

Full text of DOI:10.1021/ja049690n

Published on Web 04/01/2004
Nature of Nonbonded Se‚‚‚O Interactions Characterized by 17O
NMR Spectroscopy and NBO and AIM Analyses
Michio Iwaoka,*,† Hiroto Komatsu, Takayuki Katsuda, and Shuji Tomoda*
Contribution from the Department of Life Sciences, Graduate School of Arts and Sciences,
The UniVersity of Tokyo, Komaba, Meguro-ku, Tokyo 153-8902, Japan
Received January 16, 2004; E-mail: miwaoka@keyaki.cc.u-tokai.ac.jp; tomoda@selen.c.u-tokyo.ac.jp
Abstract: To investigate the nature of nonbonded Se‚‚‚O interactions, three series of 2-substituted
benzeneselenenyl derivatives [2-(CHO)C6H4SeX (1), 2-(CH2OH)C6H4SeX, (2), 2-(CH2OiPr)C6H4SeX (3);
X ) Cl, Br, CN, SPh, SeAr, Me] were synthesized. The 17O NMR absorption observed for 17O-enriched
aldehydes 1 appeared upfield relative to benzaldehyde (PhCHO), while the opposite downfield shifts relative
to benzyl alcohol (PhCH2OH) were observed for 17O-enriched alcohols 2 and ethers 3. The magnitude of
both the upfield and the downfield shifts became larger as the electron-withdrawing ability of a substituent
X increased. Quantum chemical calculations at the B3LYP level revealed that for all model compounds
the most stable conformer has an intramolecular nonbonded Se‚‚‚O interaction. Thus, the relative 17O
NMR chemical shifts (∆δO) for 1-3 would reflect the strengths of the Se‚‚‚O interactions. The natural bond
orbital (NBO) analysis demonstrated that the stabilization energy due to an nO f σ*Se-X orbital interaction
(ESe‚‚‚O) correlates with the Se‚‚‚O atomic distance on a single curve irrespective of the type of the O atom.
On the other hand, the atoms in molecules (AIM) analysis showed that the nonbonded Se‚‚‚O interactions
can be characterized by the presence of a bond critical point, the total energy density (HSe‚‚‚O) of which
decreases with strengthening of the interaction. The results suggested that Se‚‚‚O interactions have a
dominant covalent character rather than an electrostatic one.
Chart 1
Introduction
Weak nonbonded interactions are useful chemical tools for
controlling stability, conformation, and assembly of molecules.1
It is, therefore, of great interest to determine the strengths and
directional propensities of such interactions at an atomic
resolution. However, detection as well as characterization of
nonbonded interactions in situ is still challenging research in
fundamental chemistry.2 In this paper, we have extensively
studied the nature of nonbonded interactions between a divalent
selenium (Se) and an oxygen (O) atom, that is, Se‚‚‚O
interactions,3 by means of experimental 17O NMR spectroscopy
and theoretical natural bond orbital (NBO)4 and atoms in
molecules (AIM)5 analyses using three series of model com-
pounds (1-3) (Chart 1).
physicochemical characterization of nonbonded interactions
involving Se [e.g., Se‚‚‚N,8 Se‚‚‚O,3 Se‚‚‚F,9 and Se‚‚‚X (X )
Cl and Br)10 interactions] in relation to their applications to
asymmetric synthesis,11,12 enzyme-mimetic reactions,13 and
protein engineering.14 The nature of Se‚‚‚O interactions has been
most frequently studied in the literature.
(7) (a) Selenium in Biology and Human Health; Burk, R. F., Ed.; Springer-
Verlag: New York, 1994. (b) Selenium. Its Molecular Biology and Role
in human Health; Hatfield, D. L., Ed.; Kluwer Academic Publishers:
Boston, 2001.
Organoselenium compounds frequently possess unique chemi-
cal reactivities6 and biological activities.7 We are interested in
(8) (a) Iwaoka, M.; Tomoda, S. Phosphorus, Sulfur Silicon Relat. Elem. 1992,
67, 125-130. (b) Iwaoka, M.; Tomoda, S. J. Am. Chem. Soc. 1996, 118,
8077-8084.
† Present address: Department of Chemistry, School of Science, Tokai
University, Kitakaname, Hiratsuka-shi, Kanagawa 259-1292, Japan.
(1) (a) Israelachvili, J. Intermolecular and Surface Forces; Academic Press:
London, 1991. (b) Molecular Interactions; Scheiner, S., Ed.; John Wiley
and Sons: Chichester, 1997.
(2) Mu¨ller-Dethlefs, K.; Hobza, P. Chem. ReV. 2000, 100, 143-167.
(3) Komatsu, H.; Iwaoka, M.; Tomoda, S. Chem. Commun. 1999, 205-206.
(4) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. ReV. 1988, 88, 899-926.
(5) (a) Bader, R. F. W. Atoms in Molecules: A Quantum Theory; Oxford
University Press: New York, 1990. (b) Popelier, P. Atoms in Molecules:
An Introduction; Pearson Education: Harlow, 2000. (c) Gillespie, R. J.;
Popelier, P. L. A. Chemical Bonding and Molecular Geometry; Oxford
University Press: New York, 2001.
(9) (a) Iwaoka, M.; Komatsu, H.; Tomoda, S. Chem. Lett. 1998, 969-970. (b)
Iwaoka, M.; Komatsu, H.; Katsuda, T.; Tomoda, S. J. Am. Chem. Soc.
2002, 124, 1902-1909.
(10) Iwaoka, M.; Katsuda, T.; Tomoda, S.; Harada, J.; Ogawa, K. Chem. Lett.
2002, 518-519.
(11) (a) Fujita, K.; Iwaoka, M.; Tomoda, S. Chem. Lett. 1994, 923-926. (b)
Fujita, K.; Murata, K.; Iwaoka, M.; Tomoda, S. Tetrahedron 1997, 53,
2029-2048.
(12) (a) Fragale, G.; Wirth, T. Eur. J. Org. Chem. 1998, 1361-1369. (b) Fragale,
G.; Neuburger, M.; Wirth, T. Chem. Commun. 1998, 1867-1868.
(13) (a) Iwaoka, M.; Tomoda, S. J. Chem. Soc., Chem. Commun. 1992, 1165-
1167. (b) Iwaoka, M.; Tomoda, S. J. Am. Chem. Soc. 1994, 116, 2557-
2561. (c) Wirth, T.; Ha¨uptli, S.; Leuenberger, M. Tetrahedron: Asymmetry
1998, 9, 547-550. (d) Mugesh, G.; Singh, H. B. Chem. Soc. ReV. 2000,
29, 347-357. (e) Mugesh, G.; Panda, A.; Singh, H. B.; Punekar, N. S.;
Butcher, R. J. J. Am. Chem. Soc. 2001, 123, 839-850. (f) Back, T. G.;
Moussa, Z. J. Am. Chem. Soc. 2002, 124, 12104-12105.
(6) (a) Organoselenium Chemistry: A Practical Approach; Back, T. G., Ed.;
Oxford University Press: New York, 1999. (b) Organoselenium Chemis-
try: Modern DeVelopments in Organic Synthesis; Wirth, T., Ed.; Springer-
Verlag: Berlin, 2000.
9
10.1021/ja049690n CCC: $27.50 © 2004 American Chemical Society
J. AM. CHEM. SOC. 2004, 126, 5309-5317
5309

A R T I C L E S
Iwaoka et al.
Baiwir et al.15 unambiguously demonstrated structural features
of Se‚‚‚O interactions by solving the solid-state molecular
structure of 2-formylbenzeneselenenyl bromide (1b), in which
the carbonyl O atom coordinates intramolecularly to the Se atom
in a short distance (2.305 Å) constituting a hypervalent linear
O‚‚‚Se-Br atomic alignment. Due to this thermodynamic
assistance, the selenenyl bromide species (ArSeBr), which is
normally highly labile, could endure during the time-consuming
X-ray analysis. Chemical properties of an intramolecular 1,4-
type Se‚‚‚O interaction were studied by Goldstein et al.:16 On
the basis of X-ray and 77Se NMR analyses of biologically active
selenazofurin, it was found that the increase of a positive charge
on Se strengthens the Se‚‚‚O interaction, suggesting a significant
contribution from the electrostatic interaction between the Se
and O atoms to the stability of the attractive Se‚‚‚O interaction.
On the other hand, Barton et al.17 investigated the intramolecular
1,5-type Se‚‚‚O interaction of selenoiminoquinones in the solid
state: On the basis of ab initio molecular orbital (MO)
calculation results, the Se‚‚‚O interaction was considered as a
3-center 4-electron bond in terms of the VSEPR model.18
Recently, Minyaev and Minkin19 carried out high-level ab
initio calculations on â-chalcogenovinylaldehydes (CHOCHd
CHXR; X ) S, Se, Te and R ) H, Cl) and showed that the
molecular structures are controlled by the intramolecular 1,5-
type O f chalcogen coordination. The result was in accord with
experimental observations.15-17 However, Mo´ et al.20 pointed
out that such chalcogen‚‚‚O interactions would compete with
the X-H‚‚‚O hydrogen bond in the case of X ) Se and R )
H. They further insisted, on the basis of the results from NBO
and AIM analyses, that the electrostatic and dative (or covalent)
components are entangled in the 1,5-type Se‚‚‚O interaction.
Thus, the nature of Se‚‚‚O interactions is also controversial from
a theoretical point of view.
Scheme 1
series of model compounds 1 and 2. Although direct nuclear
spin coupling (J) was not observed between the 77Se and 17O
nuclei because of the quadrupolar nature of an 17O nucleus, both
77Se and 17O NMR chemical shifts (δSe and δO) must be
promising candidates as a probe for the nature of Se‚‚‚O
interactions. Herein, to demonstrate that the δO value can be
applied as a good experimental measure for the strength of the
Se‚‚‚O interaction, we have analyzed quantitative behaviors of
the δO values observed for 1-3 by fitting them to the
theoretically available parameter, that is, the atomic distance
between the Se and O atoms (rSe‚‚‚O). In addition, the second-
order perturbation energy due to the nO f σ*Se-X orbital
interaction (ESe‚‚‚O) obtained by NBO analysis4 and the total
energy density of the bond critical point (HSe‚‚‚O) obtained by
AIM analysis5 have been analyzed to investigate the stabilization
mechanism. The results suggested that the Se‚‚‚O interactions
of 1-3 have a dominant covalent character rather than an
electrostatic one.
Results and Discussion
Synthesis of Model Compounds. 17O-labeled 1-3 were
synthesized by using ca. 22% 17O-enriched water (H217O) as
the isotope source. According to Scheme 1, alcohol 2e8a was
oxidized to aldehyde 1e, which subsequently allowed for an
isotopic oxygen exchange reaction in the presence of an acid
catalyst. Resulting 1e* (i.e., 1e enriched with 17O) was reduced
to 2e*. 17O-enriched 3e* was easily obtained from 2e* by
alkylation of the alcohol and diselenide groups to afford 4* and
the subsequent redox conversion to recover a diselenide linkage.
Diselenides 1e*, 2e*, and 3e* were transformed to the corre-
sponding selenenyl chlorides (a, X ) Cl), selenenyl bromides
(b, X ) Br), selenenyl cyanates (c, X ) CN), selenenyl sulfides
(d, X ) SPh), and methyl selenides (f, X ) Me) by applying
common procedures.8,9 All compounds were obtained in spec-
trally pure form except for 1d*, 2d*, and 3d*, which slowly
disproportionated into the corresponding diselenides (1e*, 2e*,
and 3e*, respectively) at room temperature.
Meanwhile, Se‚‚‚O interactions have been successfully ap-
plied to asymmetric oxyselenenylation reactions of olefins using
chiral selenium reagents.12 The achievement of efficient asym-
metric induction was theoretically explained by assuming a
credit of intramolecular Se‚‚‚O interactions.21 However, it is
practically not easy to characterize Se‚‚‚O interactions in the
reaction solution. Therefore, the development of a simple
experimental method to diagnose the chemical nature and the
strength of Se‚‚‚O interactions in situ would be useful for the
design of more effective chiral selenium reagents or other
functional selenium compounds.
In our previous report on nonbonded Se‚‚‚O interactions,3
77Se and 17O NMR chemical shifts (δSe and δO) were shown to
be sensitive to the presence of Se‚‚‚O interactions by using a
17O and 77Se NMR Chemical Shifts. The 17O and 77Se NMR
chemical shifts (δO and δSe) observed for 1a-f(*), 2a-f(*),
and 3a-f(*) are listed in Table 1. On the basis of a similar
discussion previously reported,3 the presence of intramolecular
Se‚‚‚O interactions for these series of model compounds can
be easily deduced from the NMR data.
First, δO values for 1 (δO ) 493.0-561.6) were shifted upfield
(∆δO ) -76 to -7) as compared to that of reference compound
benzaldehyde (PhCHO, δO ) 569),22 whereas δO values for 2
(δO ) 24.9-10.6) and those for 3 (δO ) 55.2-39.0) were
(14) (a) Iwaoka, M.; Takemoto, S.; Okada, M.; Tomoda, S. Chem. Lett. 2001,
132-133. (b) Iwaoka, M.; Takemoto, S.; Okada, M.; Tomoda, S. Bull.
Chem. Soc. Jpn. 2002, 75, 1611-1625. (c) Iwaoka, M.; Takemoto, S.;
Tomoda, S. J. Am. Chem. Soc. 2002, 124, 10613-10620.
(15) Baiwir, M.; Llabres, G.; Dideberg, O.; Dupont, L.; Piette, J. L. Acta
Crystallogr. 1975, B31, 2188-2191.
(16) (a) Goldstein, B. M.; Kennedy, S. D.; Hennen, W. J. J. Am. Chem. Soc.
1990, 112, 8265-8268. (b) Burling, F. T.; Goldstein, B. M. J. Am. Chem.
Soc. 1992, 114, 2313-2320.
(17) Barton, D. H. R.; Hall, M. B.; Lin, Z. Y.; Parekh, S. I.; Reibenspies, J. J.
Am. Chem. Soc. 1993, 115, 5056-5059.
(18) (a) Gillespie, R. J.; Robinson, E. A. Angew. Chem., Int. Ed. Engl. 1996,
35, 495-514. (b) Gillespie, R. J. Coord. Chem. ReV. 2000, 197, 51-69.
(19) Minyaev, R. M.; Minkin, V. I. Can. J. Chem. 1998, 76, 776-788.
(20) (a) Sanz, P.; Ya´n˜ez, M.; Mo´, O. J. Phys. Chem. A 2002, 106, 4661-4668.
(b) Sanz, P.; Mo´, O.; Ya´n˜ez, M. Phys. Chem. Chem. Phys. 2003, 5, 2942-
2947.
(22) Boykin, D. W.; Baumstark, A. L. In 17O NMR Spectroscopy in Organic
Chemistry; Boykin, D. W., Ed.; CRC Press: Boca Raton, FL, 1991; pp
205-231.
(21) Spichty, M.; Fragale, G.; Wirth, T. J. Am. Chem. Soc. 2000, 122, 10914-
10916.
9
5310 J. AM. CHEM. SOC. VOL. 126, NO. 16, 2004

Nature of Nonbonded Se‚‚‚O Interactions
A R T I C L E S
Table 1. 77Se and 17O NMR Chemical Shifts (δSe and δO) for
1-3a
1 (CHO)
2 (CH OH)
3 (CH OiPr)
2
2
b
b
b
b
X
δSe
δO
δSe
δO
δSe
δO
a
b
c
d
e
f
Cl
Br
CN
SPh
SeAr
Me
1114.1
1029.5
426.7
621.7
458.5
259.5
493.0
515.9
548.0
556.6
559.3
561.6
987.1
839.5
314.5
501.8
433.2
157.2
24.9
23.2
22.0
16.0
13.3
10.6
986.5
857.8
315.1
503.8
412.0
166.6
54.8
55.2
44.6
44.4
39.0
39.0
a 77Se NMR spectra were measured at 95.35 MHz in CDCl3 at 298 K
with Me2Se as an external standard. 17O NMR spectra were measured for
the corresponding 17O-enriched compounds at 67.70 MHz in CDCl3 at 298
K with D2O as an external standard. b Data from ref 3.
Chart 2
shifted downfield (∆δO ) 24.2-9.9 and 54.5-38.3, respec-
tively) as compared to that of reference compound benzyl
alcohol (PhCH2OH, δO ) 0.7).23 The observed opposite shifts
were reasonably due to the different types of hybrid orbitals
for the O atom of 1 (sp2) and those of 2 and 3 (sp3).24 Second,
the magnitudes of both the upfield and the downfield shifts
became larger as the electron-withdrawing ability of a substituent
X increased, that is, a (Cl) > b (Br) > c (CN) > d (SPh) > e
(SeAr) > f (Me), except for the case of 3a (∆δO ) 54.1) and
3b (∆δO ) 54.5): The monotonic downfield shift of δO was
observed for the series of compound 1 on going from 1a to 1f
(δO ) 493.0 f 561.6), while the opposite trend was apparent
for the series of 2 (δO ) 24.9 f 10.6) and 3 (δO ) 55.2 f
39.0). Third, the 77Se NMR absorptions (δSe) for 1a-f were
significantly shifted downfield from those for the corresponding
2a-f and 3a-f, suggesting the presence of strong magnetic
anisotropic effects on the Se atoms of series 1a-f from the
intramolecular carbonyl group that must locate in close proxim-
ity to the Se atom.
Figure 1. The most stable structures for 1a-f obtained by quantum
chemical calculation at the B3LYP/631H//B3LYP/631H level. The pertinent
bond lengths are given in angstroms.
Thus, the NMR data in Table 1 allowed us to suppose that
the major conformer in solution for three series of model
compounds (1a-f, 2a-f, and 3a-f) should be the one that has
a close atomic contact between the Se and O atoms, that is, an
intramolecular Se‚‚‚O interaction.
Quantum Chemical Calculation. To assign the most stable
molecular structure for each model compound (1a-f, 2a-f, and
3a-f) by theoretical calculation, a systematic conformer search,
in which all possible values were applied for each changeable
dihedral angle, was carried out at the HF/631H level (see
Experimental Section for the abbreviation). The obtained stable
structures were subsequently optimized without restrictions at
the B3LYP/631H level. Simplified structures were employed
for some model compounds to save the computation time.
Isopropyl groups of 3a-f were treated as methyl groups in 5a-f
(Chart 2). Phenylthio substituents (SPh) of 1d, 2d, and 3d were
altered to methylthio groups (SMe) in 1d′, 2d′, and 5d′.
Figure 2. The most stable structures for 2a-f obtained by quantum
chemical calculation at the B3LYP/631H//B3LYP/631H level. The pertinent
bond lengths are given in angstroms.
Similarly, arylseleno substituents (SeAr) of 1e, 2e, and 3e were
altered to methylseleno groups (SeMe) in 1e′, 2e′, and 5e′. The
simplification should not significantly affect the calculation
results.
The most stable molecular structures obtained for 1a-f, 2a-
f, and 5a-f in vacuo at the B3LYP/631H level are shown in
Figures 1-3, respectively. For each model compound, the
conformer with an intramolecular Se‚‚‚O interaction was found
to be most stable. The relative stability to other stable conform-
(23) Balakrishnan, P.; Baumstark, A. L.; Boykin, D. W. Tetrahedron Lett. 1984,
25, 169-172.
(24) (a) Be´raldin, M. T.; Vauthier, E.; Flisza´r, S. Can. J. Chem. 1982, 60, 106-
110. (b) Jaccard, G.; Carrupt, P. A.; Lauterwein, J. Magn. Reson. Chem.
1988, 26, 239-244.
9
J. AM. CHEM. SOC. VOL. 126, NO. 16, 2004 5311

A R T I C L E S
Iwaoka et al.
Table 3. Summary of Quantum Chemical Calculations on 1a-f,
2a-f, and 5a-f and the Natural Bond Orbital (NBO) Analysisa
b
c
d
d
e
rSe‚‚‚O θO‚‚‚Se-X ωAr-Se
(Å) (deg) (deg)
ωAr-C
(deg)
qSe
(e)
qO
(e)
ESe‚‚‚O
(kcal/mol)
comp
X
1a Cl
1b Br
1c CN
1d′ SMe 2.644 176.4 178.5
1e′ SeMe 2.661 177.4 178.5
2.305 176.4 180.0
2.336 177.6 180.0
2.594 171.9 180.0
0.0 0.630 -0.551 36.07
0.0 0.538 -0.548 33.80
0.0 0.570 -0.542 11.33
0.6 0.390 -0.537 10.72
0.5 0.302 -0.530 10.66
1f
Me
2.758 173.9 180.0
0.0 0.487 -0.536
6.19
2a Cl
2b Br
2.567 175.2 161.3 39.2 0.485 -0.739 14.80
2.604 175.7 159.7 40.8 0.401 -0.738 13.61
2c
CN
2.787 170.9 158.7 46.3 0.500 -0.753
5.95
4.71
4.46
2.57
2d′ SMe 2.873 174.5 157.6 49.4 0.308 -0.751
2e′ SeMe 2.901 175.1 156.9 50.8 0.222 -0.748
2f
Me
3.005 171.0 156.2 54.9 0.402 -0.751
5a Cl
5b Br
2.555 175.5 161.5 39.5 0.490 -0.566 14.38
2.596 175.9 159.9 41.2 0.404 -0.565 13.17
5c
CN
2.795 170.9 158.7 47.2 0.498 -0.578
4.86
4.21
3.87
2.26
5d′ SMe 2.876 175.2 159.3 49.5 0.310 -0.571
5e′ SeMe 2.912 175.2 157.4 51.5 0.218 -0.569
5f
Me
3.007 171.3 156.9 54.4 0.400 -0.571
a Selected structural parameters are shown for the most stable conformers
(see Figures 1-3) obtained at the B3LYP/631H//B3LYP/631H level. NBO
analysis was performed at the same calculation level. b The dihedral angle
of the X-Se-C-CC linkage. c The dihedral angle of the O-C-C-CSe
linkage. d The atomic charge of the selenium or oxygen atom determined
by natural population analysis. e The orbital interaction energy between the
oxygen lone pairs (nO) and the antibonding orbital (σ*Se-X) determined by
NBO second-order perturbation analysis.
Figure 3. The most stable structures for 5a-f obtained by quantum
chemical calculation at the B3LYP/631H//B3LYP/631H level. The pertinent
bond lengths are given in angstroms.
Table 2. Structural Parameters for 2-Formylbenzeneselenenyl
Bromide (1b) Obtained by Quantum Chemical Calculation in
Comparison to Those Determined by X-ray Crystallographic
Analysis
significantly, shorter in the solid state than those in the calculated
structure. The discrepancy may be due to intermolecular
interactions in the solid state, as previously explained for the
shrinking of hypervalent Br-Se-Br and Cl-Se-Cl frag-
ments.25 The potential surface of the hypervalent Br-Se‚‚‚O
fragment of 1b should be so shallow, thereby so sensitive to
the surrounding environment, that the fragment may be slightly
shrunk in the solid state due to the crystal packing force.
Table 3 summarizes the structural parameters of the most
stable conformers that were assigned for 1a-f (Figure 1), 2a-f
(Figure 2), and 5a-f (Figure 3) by quantum chemical calcula-
tion. Compounds 1a-f possessed a planar structure according
to the values of dihedral angles for the Se-X and C-O bonds
atomic distance
B3LYP (Å)a
X-ray (Å)b
difference (Å)
Se-Br
Se‚‚‚O
Se-C
CdO
C-C
C-C
2.448
2.336
1.919
1.241
1.437
1.401c
2.403(4)
2.305(19)
1.876(19)
1.248(33)
1.420(36)
1.400c
0.045
0.031
0.043
-0.007
0.017
0.001
angle
B3LYP (deg)a
X-ray (deg)b
difference (deg)
Br-Se‚‚‚O
Br-Se-C
O‚‚‚Se-C
CdO‚‚‚Se
OdC-C
C-C-CSe
C-C-CH
Se-C-CC
Se-C-CH
C-C-C
177.6
98.7
79.0
107.1
120.8
117.1
122.1
116.1
125.4
120.0c
176.8
0.8
0.7
-0.1
1
-2
4
98.0(6)
79.1(8)
106(2)
123(2)
113(2)
122(2)
118(1)
126(2)
120c
with respect to the aromatic ring (i.e., ωAr-Se and ωAr-C
,
respectively), while distorted structures were obtained for 2a-f
and 5a-f as seen in the values of ωAr-Se (156.2-161.5°) and
ωAr-C (39.2-54.9°). The structural parameters for 2a-f were
very similar to those for 5a-f, suggesting that the displacement
of the hydroxy groups of 2a-f with alkoxy groups in 5a-f
would make only marginal effects on the conformation. The
apparently short atomic distances between Se and O atoms (rSe‚
0
-2
-1
0
a Optimized at the B3LYP/631H level. b Data from ref 15. c Averaged
values.
‚‚ ) relative to the sum of the van der Waals radii [Vdw(Se) +
O
Vdw(O) ) 1.90 + 1.52 ) 3.42 Å]26 as well as the almost linear
O‚‚‚Se-X angles (θO‚‚‚Se-X) clearly show the presence of a
hypervalent Se‚‚‚O interaction for all model compounds. It
should be noted that rSe‚‚‚O was 0.1-0.3 Å longer when
calculated at the HF/631H level.3 Because the electron correla-
tion is more sufficiently included in the B3LYP method than
in the HF method, the observed shortening of rSe‚‚‚O at the
B3LYP/631H level (Table 3) suggested the contribution from
the electron correlation to the stability of the Se‚‚‚O interactions.
ers that do not have an intramolecular Se‚‚‚O interaction ranged
from 3.0 to 11.6 kcal/mol for 1a-f, 0.4 to 3.5 kcal/mol for
2a-f, and 0.4 to 3.0 kcal/mol for 5a-f.
Accuracy of the calculation results was confirmed by the good
agreement between the molecular structure of 1b obtained in
silico and that determined by X-ray analysis.15 Table 2 compares
the structural parameters of 1b. All atomic distances and angles
of 1b in the solid state are reasonably reproduced by the
theoretical calculation at the B3LYP/631H level within the
experimental errors, except that the atomic distances around the
Se atom (i.e., Se-Br, Se‚‚‚O, and Se-C) are slightly, but
(25) Iwaoka, M.; Komatsu, H.; Tomoda, S. J. Organomet. Chem. 2000, 611,
164-171.
(26) Bondi, A. J. Phys. Chem. 1964, 68, 441-451.
9
5312 J. AM. CHEM. SOC. VOL. 126, NO. 16, 2004

Nature of Nonbonded Se‚‚‚O Interactions
A R T I C L E S
Figure 4. Correlation between the 17O NMR chemical shifts (δO) and the
Se‚‚‚O atomic distances (rSe‚‚‚O) calculated at the B3LYP/631H//B3LYP/
631H level. (a) Correlation plots for 1a-f (CHO). The dashed line indicates
the 17O NMR chemical shift for PhCHO (δO ) 569).22 (b) Correlation plots
Figure 5. Correlation plots of the NBO second-order perturbation energies
(ESe‚‚‚O) versus the Se‚‚‚O atomic distances (rSe‚‚‚O) calculated at the B3LYP/
631H//B3LYP/631H level. The solid curve is drawn for convenience.
for 2a-f (CH2OH) and 3a-f (CH2OiPr). The dashed line indicates the 17
O
NMR chemical shift for PhCH2OH (δO ) 0.7).23 The Se‚‚‚O atomic
distances (rSe‚‚‚O) calculated for 5a-f (CH2OMe) are applied for 3a-f.
NBO Analysis. To evaluate the contribution from the orbital
interaction between Se and O atoms to the stability of Se‚‚‚O
interactions, the NBO second-order perturbation analysis4 was
applied for the most stable conformers of 1a-f, 2a-f, and 5a-f
at the B3LYP/631H level. The obtained stabilization energies
due to the nO f σ*Se-X orbital interactions (ESe‚‚‚O) are listed
in the last column of Table 3. Because the O atom has two
lone pairs of electrons, the ESe‚‚‚O values are indicated in the
sum of the two possible nO f σ*Se-X orbital interactions.
In the NBO analysis, natural bond orbitals are first defined
for each covalent bond, lone pair, and antibonding orbital by
using the molecular orbitals obtained by quantum chemical
calculation, and the orbital interaction energies (i.e., the NBO
second-order perturbation energies) are subsequently analyzed
for all possible combinations of the two natural bond orbitals.
Like Se‚‚‚N8 and Se‚‚‚F9 interactions, the orbital interactions
between the O lone pairs (nO) and the antibonding orbital of
the Se-X bond (σ*Se-X) were found to contribute significantly
to stabilization of the Se‚‚‚O interactions: the ESe‚‚‚O values
ranged from 6.19 to 36.07 kcal/mol for 1a-f, 2.57 to 14.80
kcal/mol for 2a-f, and 2.26 to 14.38 kcal/mol for 5a-f.
The nO f σ*Se-X orbital interaction energies (ESe‚‚‚O) decrease
on going from a (X ) Cl) to f (X ) Me) for all series. On the
other hand, for the compounds with the same substituent X, 1
(CHO) has a much larger value of ESe‚‚‚O than 2 (CH2OH), and
2 has approximately the same value of ESe‚‚‚O as 5 (CH2OMe).
The trends indicated the importance of the electrophilicity of
the Se-X bond and the nucleophilicity of the O atom for the
Se‚‚‚O interactions. Because the observed trends are in complete
agreement with the tendency of the Se‚‚‚O atomic distance (rSe‚
Because the Se‚‚‚O atomic distance (rSe‚‚‚O) is a simple and
unambiguous measure for the strength of Se‚‚‚O interactions,
their correlation to the 17O NMR chemical shifts (δO) observed
for 1a-f, 2a-f, and 3a-f is plotted in Figure 4. It is seen that
the shorter the Se‚‚‚O atomic distance, the larger the upfield
shift of 17O NMR for 1a-f relative to PhCHO (a dashed line
in Figure 4a) and the larger the downfield shift of 17O NMR
for 2a-f and 3a-f relative to PhCH2OH (a dashed line in Figure
4b). The monotonic correlations demonstrated that the major
conformer for 1a-f, 2a-f, and 3a-f in the NMR solvent
(CDCl3) possesses an Se‚‚‚O interaction and also that the
strength of the Se‚‚‚O interaction decreases on going from a
(X ) Cl) to f (X ) Me) for all series of compounds.
Conformations of 1a-f, 2a-f, and 3a-f in solution should
be affected by the solvent effect, which generally tends to
stabilize a more polar conformer than a less polar one due to
more efficient solvation around the exposed polar surface. This
implies that the conformer with an intramolecular Se‚‚‚O
interaction, which has the polar surface less exposed than other
open conformers, would be destabilized in solution to some
extent. According to the correlations shown in Figure 4,
however, all model compounds should keep an intramolecular
Se‚‚‚O interaction in the NMR solution. This in turn indicates
that the solvent effects on the conformations of 1a-f, 2a-f,
and 3a-f would be weak.
The atomic charges of Se and O atoms for 1a-f, 2a-f, and
5a-f (qSe and qO in Table 3) suggest the significance of the
electrostatic interaction between a positively charged Se atom
and a negatively charged O atom for the stability of the Se‚‚‚O
interactions. However, the values of qSe and qO do not seem to
correlate well with the strength (i.e., the trends of the 17O NMR
and the Se‚‚‚O atomic distance), except for the monotonic
tendency that the qO values calculated for 1a-f become less
negative on going from 1a to 1f. In our previous calculation at
the HF/631H level,3 it was found that the higher the charge of
the O atom, the larger the upfield shift of 17O NMR for 1a-f
and the larger the downfield shift of 17O NMR for 2a-f. The
calculation results at the B3LYP/631H level (Table 3) took over
the trend observed at the lower calculation level (i.e., the HF/
631H level) for 1a-f, but not for 2a-f. The considerations
suggested that the nature of the Se‚‚‚O interaction cannot solely
be explained by the electrostatic interaction between Se and O
atoms.
‚‚ ) calculated for 1a-f, 2a-f, and 5a-f, the nO f σ*Se-X
O
orbital interaction mechanism reasonably explains the tendency
of the strength of the Se‚‚‚O interactions. The nice correlation
between ESe‚‚‚O and rSe‚‚‚O (Figure 5) strongly suggests that the
nO f σ*Se-X orbital interaction is a major factor for the stability
of the Se‚‚‚O interactions. It is noteworthy that ESe‚‚‚O and rSe‚
‚‚O exhibit a single correlation curve irrespective of the type of
the O atom (i.e., whether it is involved in CHO, CH2OH, or
CH2OR groups), while such a simple correlation is not present
between the atomic charges of Se and O atoms (qSe and qO)
and their atomic distance (rSe‚‚‚O).
AIM Analysis. To investigate the nature of Se‚‚‚O interaction
from another point of view, atoms in molecules (AIM) analysis5
was performed by using the calculation results at the B3LYP/
631H level. In the AIM analysis, topological properties of the
9
J. AM. CHEM. SOC. VOL. 126, NO. 16, 2004 5313

A R T I C L E S
Iwaoka et al.
Table 4. Bond Critical Point (BCP) Properties of Se‚‚‚O
Interactions for 1a-f, 2a-f, and 5a-fa
b
2
c
d
FSe‚‚‚O
(ea0-3
∇ FSe‚‚‚O
(ea0-5
H
Se‚‚‚O
comp
X
)
)
(ea0-4
)
1a
1b
1c
1d′
1e′
1f
Cl
Br
CN
SMe
SeMe
Me
0.053
0.050
0.030
0.028
0.027
0.022
0.130
0.124
0.084
0.077
0.075
0.065
-0.0053
-0.0045
-0.0002
-0.0002
-0.0002
0.0002
2a
2b
2c
2d′
2e′
2f
Cl
Br
CN
SMe
SeMe
Me
0.032
0.030
0.021
0.018
0.018
0.014
0.084
0.079
0.060
0.051
0.049
0.043
-0.0012
-0.0011
0.0001
0.0000
0.0000
0.0003
Figure 6. Correlation plots of the AIM total energy densities (HSe‚‚‚O) versus
the Se‚‚‚O atomic distances (rSe‚‚‚O) calculated at the B3LYP/631H//B3LYP/
631H level. The solid curve is drawn for convenience.
5a
5b
5c
5d′
5e′
5f
Cl
Br
CN
SMe
SeMe
Me
0.032
0.030
0.020
0.018
0.017
0.014
0.085
0.079
0.058
0.050
0.048
0.042
-0.0013
-0.0011
0.0002
0.0001
0.0001
0.0004
However, it was previously pointed out that the total energy
density (H) at a BCP, instead of the Laplacian, is a more
appropriate index to approach better understanding of weak
nonbonded interactions.27 A total energy density (H) is given
as a sum of two energy density terms; H ) G + V, where G is
the electronic kinetic energy density and V is the electronic
potential energy density. When the value of H at a BCP is
positive (i.e., G > -V), accumulation of electrons at the BCP
is destabilizing. When the value of H at a BCP is negative (i.e.,
G < -V), accumulation of electrons at the BCP is stabilizing.
Thus, within the framework of the AIM theory, the sign of H
at the BCP assigns whether the interaction is electrostatic
dominant (H > 0) or covalent dominant (H < 0). It should be
noted that a covalent interaction in the AIM theory relates to
an orbital interaction in the NBO analysis.
a AIM analysis was performed for the most stable conformers (see Figures
1-3) obtained at the B3LYP/631H//B3LYP/631H level. b The electron
density at the BCP. c The Laplacian of the electron density at the BCP.
d The total energy density at the BCP.
electron density are analyzed to define the bond path between
bonding (or interacting) atoms. Chemical bondings can be
characterized by a so-called bond critical point (BCP), where
the electron density becomes a minimum value along the bond
path. For all Se‚‚‚O interactions of model compounds (in the
molecular structures shown in Figures 1-3), the BCP could be
located. Table 4 summarizes the BCP properties of Se‚‚‚O
interactions for 1a-f, 2a-f, and 5a-f.
The electron density (F) at a BCP correlates with the strength
of an atomic interaction. The values of F obtained for the Se‚
‚‚O interactions of 1a-f, 2a-f, and 5a-f (FSe‚‚‚O) ranged from
0.014 to 0.053 ea0-3, which are remarkably lower than normal
According to the values of HSe‚‚‚O listed in Table 4, it seems
that the Se‚‚‚O interactions have a dominant covalent character,
although when the Se‚‚‚O interaction is weak, the value of HSe‚
‚‚ becomes slightly positive. Figure 6 shows the correlation
O
plots between the total electron density at the BCP (HSe‚‚‚O) and
the atomic distance between the Se and O atoms (rSe‚‚‚O). It is
seen that HSe‚‚‚O becomes more negative with a decrease in the
Se‚‚‚O atomic distance (i.e., with strengthening of the Se‚‚‚O
interaction). The correlation, which is independent of the type
of the O atom, supports the consideration that the Se‚‚‚O
interactions have a distinctly covalent nature rather than an
electrostatic one.
covalent bonds (FC-C ≈ 0.24 ea0
)
-3 5b but are significantly higher
than the practical boundary of a molecule (F ≈ 0.001 ea0-3).
The range was similar to that for neutral hydrogen bonds
(FH-bond ≈ 0.002-0.04 ea0-3),5b suggesting that the strength of
the Se‚‚‚O interactions competes with that of hydrogen bonds.
The tendency that FSe‚‚‚O decreases in the order of a (X ) Cl)
> b (X ) Br) > c (X ) CN) > d′ (X ) SMe) > e′ (X ) SeMe)
> f (X ) Me) is in full accordance with the tendency of ESe‚‚‚O
obtained by NBO analysis as well as rSe‚‚‚O obtained by quantum
chemical calculation (see Table 3).
Conclusions
As to the nature of Se‚‚‚O interactions, both covalent and
electrostatic characters should be important for the stability.
However, NBO (i.e., ESe‚‚‚O) and AIM (i.e., HSe‚‚‚O) analyses
presented in this paper strongly suggested that the nO f σ*Se-X
orbital interactions are dominantly important, although the
Laplacian of the electron density at the bond critical point
2
The Laplacian (∇ FSe‚‚‚O) represents the curvature of the
electron density in three-dimensional space at the BCP of the
2
Se‚‚‚O interaction. In general, a negative value of ∇ F indicates
that the electron density is locally concentrated, while a positive
2
value of ∇ F means that the electron density is locally depleted.
2
2
(∇ FSe‚‚‚O) suggested dominance of the electrostatic interaction
Therefore, the sign of ∇ F at a BCP is considered to relate
between a positively charged Se and a negatively charged O
atom. Because the total energy density (HSe‚‚‚O) may be a better
probe for characterization of weak interactions than the Lapla-
cian,27 it must be reasonable to conclude that the Se‚‚‚O
interactions for 1a-f, 2a-f, and 3a-f have a dominant covalent
character rather than an electrostatic one. This was also
whether the atomic interaction possesses a dominant character
2
of the shared electron (covalent) interactions (∇ F < 0) or the
2
closed-shell (electrostatic) interactions (∇ F > 0). The values
2
of ∇ FSe‚‚‚O listed in Table 4 are all positive, suggesting that
the Se‚‚‚O interactions have a dominant electrostatic character.
Moreover, their magnitude decreases on going from a (X )
Cl) to f (X ) Me). The results are in conflict with the earlier
discussion based on the NBO analysis (i.e., the tendency of ESe‚
(27) (a) Koch, W.; Frenking, G.; Gauss, J.; Cremer, D.; Collins, J. R. J. Am.
Chem. Soc. 1987, 109, 5917-5934. (b) Rozas, I.; Alkorta, I.; Elguero, J.
J. Am. Chem. Soc. 2000, 122, 11154-11161.
‚‚ ).
O
9
5314 J. AM. CHEM. SOC. VOL. 126, NO. 16, 2004

Nature of Nonbonded Se‚‚‚O Interactions
A R T I C L E S
17O-Labeled Bis(2-formylphenyl) Diselenide (1e*). To a 45.6 mM
p-dioxane solution of 1e (60 mL, 2.74 mmol) were added a 20:1 v/v
mixture (1.04 mL) of 17O-labeled hydrogen oxide (21.89 at. % H217O;
CDN Isotope, Inc.) and concentrated hydrochloric acid (36%, 11.85
M) under nitrogen atmosphere. After being stirred for 17 h at 30 °C,
the reaction mixture was evaporated in vacuo. 17O-labeled 1e* was
obtained in spectrally pure form (1.00 g, 99% yield). Spectral data for
1e*: 17O NMR δ 559.3 (broad s).
supported by the poor correlation observed between the tendency
of the atomic charges of the Se and O atoms (qSe and qO,
respectively) and the distance between the Se and O atoms (see
Table 3). Meanwhile, the electron correlation was also suggested
to be important for the Se‚‚‚O interactions in addition to the
covalent and electrostatic factors, according to significant
elongation of the Se‚‚‚O atomic distance in the HF structures,3
which do not include the electron correlation sufficiently.
2-Formylbenzeneselenenyl Chloride (1a). Compound 1e (25.8 mg,
0.07 mmol) was dissolved in dry THF (3 mL) under nitrogen. To the
solution was added sulfuryl chloride (6.8 µL, 0.08 mmol). After being
stirred for 1 h at room temperature, the reaction mixture was
concentrated under reduced pressure. The residual brown oil was pure
The above considerations are in accord with the nature of
Se‚‚‚N8 and Se‚‚‚F9 interactions. Nonbonded Se‚‚‚N and Se‚‚
‚F interactions have been clearly characterized by the observa-
tion of the nuclear spin-spin coupling between the interacting
atoms, which gave clear evidence to the significance of the
orbital interactions. As for Se‚‚‚F interactions, the importance
of the electron correlation has been additionally suggested for
the stability. Because oxygen locates between nitrogen and
fluorine in the periodic table of elements, it is quite consistent
that Se‚‚‚O interactions have an intermediate chemical character
between Se‚‚‚N and Se‚‚‚F interactions.
From a practical point of view, it is important to develop a
simple experimental method to determine the strength of Se‚‚
‚O interactions. The correlation shown in Figure 4 demonstrated
that the 17O NMR chemical shift (δO) is useful as the probe,
but it is not always easy to observe 17O NMR in laboratories
unless the sample compound has been labeled with an 17O
nucleus. On the other hand, the correlations shown in Figures
5 and 6 strongly suggested that the Se‚‚‚O interactions involving
various types of an O atom can be treated as one class: the
single correlation between the strength and the covalent character
covers the range of the three types of Se‚‚‚O interactions. The
discovered correlations would be useful when the strength of
the Se‚‚‚O interaction is estimated from the Se‚‚‚O atomic
distance. Another possibility to apply the 77Se NMR chemical
shift (δSe) as a probe for the strength of Se‚‚‚O interactions will
be discussed in due course.
1
1a according to the H, 13C, and 77Se NMR spectra. Spectral data for
1a: 1H NMR δ 10.27 (s, 1H), 8.24 (d, J ) 8.0 Hz, 1H), 8.09 (dd, J )
8.0 and 1.5 Hz, 1H), 7.78 (td, J ) 8.0 and 1.5 Hz, 1H), 7.51 (t, J )
8.0 Hz, 1H); 13C NMR δ 192.5, 148.7, 135.7, 133.6, 132.7, 127.4, 126.1;
77Se NMR δ 1114.1 (lit. 109728). A similar procedure was applied to
1e* for the synthesis of 17O-labeled 1a (1a*). Spectral data for 1a*:
17O NMR δ 493.0 (broad s).
2-Formylbenzeneselenenyl Bromide (1b). Compound 1e (27.9 mg,
0.08 mmol) was dissolved in CH2Cl2 (3 mL) under nitrogen. To the
solution was added bromine (4.0 µL, 0.08 mmol). After 15 min, the
reaction mixture was evaporated. The residual brown solid was pure
1
1b according to the H, 13C, and 77Se NMR spectra. Spectral data for
1b: 1H NMR δ 10.12 (s, 1H), 8.19 (d, J ) 8.4 Hz, 1H), 8.01 (dd, J )
8.4 and 1.5 Hz, 1H), 7.72 (td, J ) 8.4 and 1.5 Hz, 1H), 7.51 (t, J )
8.4 Hz, 1H); 13C NMR δ 192.4, 144.0, 135.6, 134.0, 133.1, 130.4, 126.2;
77Se NMR δ 1029.5 (lit. 101928). A similar procedure was applied to
1e* for the synthesis of 17O-labeled 1b (1b*). Spectral data for 1b*:
17O NMR δ 515.9 (broad s).
2-Formylbenzeneselenenyl Cyanate (1c). Compound 1a, prepared
from 1e (33.7 mg, 0.09 mmol), was dissolved in dry THF (3.7 mL)
under nitrogen. To the solution was added cyanotrimethylsilane (120
µL, 0.91 mmol). After 2 h, the reaction mixture was concentrated under
reduced pressure. Product 1c was obtained in 84% yield (32.3 mg) as
a colorless oil after purification of the crude product by silica gel column
chromatography (CH2Cl2). Spectral data for 1c: 1H NMR δ 10.07 (s,
1H), 8.03 (d, J ) 8.2 Hz, 1H), 7.93 (dd, J ) 8.2 and 1.4 Hz, 1H), 7.68
(td, J ) 8.2 and 1.4 Hz, 1H), 7.60 (t, J ) 8.2 Hz, 1H); 13C NMR δ
192.8, 135.7, 135.6, 132.6, 130.3, 129.8, 127.8, 104.2; 77Se NMR δ
426.7 (lit. 42328). Anal. Calcd for C8H5NOSe: C, 45.74; H, 2.40; N,
6.67. Found: C, 45.77; H, 2.51; N 6.79. A similar procedure was
applied to 1e* for the synthesis of 17O-labeled 1c (1c*). Spectral data
for 1c*: 17O NMR δ 548.0 (broad s).
Experimental Section
General Procedures. Commercially available organic and inorganic
reagents were used without further purification. Tetrahydrofuran (THF)
was dried over sodium wire and was distilled under nitrogen. Dichlo-
romethane (CH2Cl2) was dried over calcium hydride and was distilled
under nitrogen before use. Methanol (MeOH) was distilled under
2-Formylbenzeneselenenyl Phenyl Sulfide (1d). Compound 1e
(54.9 mg, 0.15 mmol) was dissolved in CH2Cl2 (6 mL). To the solution
were added pyridine (3 drops) and benzenethiol (76.5 µL, 0.75 mmol).
After 3 d, the reaction mixture was concentrated under reduced pressure
to afford a yellow oil, which contained 1d, 1e, and diphenyl disulfide.
1
nitrogen. Other organic solvents were used without purification. H
(500 MHz), 13C (125.65 MHz), 17O (67.70 MHz), and 77Se (95.35 MHz)
NMR spectra were measured at 298 K on a JEOL R500 spectrometer.
The sample was dissolved in CDCl3 containing tetramethylsilane as
1
an internal standard for H and 13C NMR. For 17O NMR, D2O (δ 0.0
1
The yield of 1d was 78% according to the integration of H NMR.
ppm) was used as an external standard. For 77Se NMR, dimethyl
Due to the disproportionation of 1d, purification could not be performed.
Spectral data for 1d: 1H NMR δ 10.12 (s, 1H), 8.21 (d, J ) 7.1 Hz,
1H), 7.87 (dd, J ) 7.1 and 1.3 Hz, 1H), 7.56 (td, J ) 7.1 and 1.3 Hz,
1H), 7.48 (dd, J ) 7.1 and 1.3 Hz, 2H), 7.43 (td, J ) 7.1 and 1.3 Hz,
1H), 7.22 (d, J ) 7.1 Hz, 2H), 7.16 (tt, J ) 7.1 and 1.3 Hz, 1H); 13C
NMR δ 193.0, 138.6, 135.9, 135.6, 134.4, 134.2, 129.1, 129.0, 127.9,
126.8, 126.2; 77Se NMR δ 621.7. A similar procedure was applied to
1e* for the synthesis of 17O-labeled 1d (1d*). Spectral data for 1d*:
17O NMR δ 556.6 (broad s).
selenide (δ 0 ppm) in CDCl3 was used as an external standard.
Bis(2-formylphenyl) Diselenide (1e). Bis[2-(hydroxymethyl)phenyl]
diselenide (2e)8a (2.06 g, 5.5 mmol) was dissolved in dimethyl sulfoxide
(6 mL) under nitrogen atmosphere. To the solution was added
trimethylsilyl chloride (3.5 mL, 27.6 mmol). After 15 min, the reaction
was worked up with aqueous sodium hydrogencarbonate, and the
mixture was extracted with CH2Cl2. The organic layer was collected,
dried over anhydrous sodium sulfate, filtrated, and then evaporated.
The resulting crude product was purified by silica gel column
chromatography (benzene) to yield 1e (1.05 g, 52%) as slightly yellow
crystals. Spectral data for 1e: 1H NMR δ 10.17 (s, 2H), 7.87-7.80
(m, 4H), 7.44-7.39 (m, 4H); 13C NMR δ 192.9, 135.9, 134.8, 134.6,
134.4, 131.1, 126.3; 77Se NMR δ 458.5. Anal. Calcd for C14H10O2Se2:
C, 45.68; H, 2.74. Found: C, 45.43; H, 2.82.
2-Formylphenyl Methyl Selenide (1f). To a dimethyl sulfoxide (0.5
mL) solution of potassium hydroxide (18.0 mg, 0.32 mmol) were added
compound 1e (30.5 mg, 0.08 mmol) and methyl iodide (21 µL, 0.34
(28) Llabre`s, G.; Baiwir, M.; Piette, J.-L.; Christiaens, L. Org. Magn. Reson.
1981, 15, 152-154.
9
J. AM. CHEM. SOC. VOL. 126, NO. 16, 2004 5315

A R T I C L E S
Iwaoka et al.
mmol) under nitrogen. After 2 h, the reaction was worked up with
water, and the mixture was extracted with CH2Cl2. Product 1f was
obtained as a colorless oil (16.2 mg, 82%) after purification of the
crude product by silica gel column chromatography (benzene). Spectral
data for 1f: 1H NMR δ 10.15 (s, 1H), 7.81 (dd, J ) 7.7 and 1.6 Hz,
1H), 7.50 (td, J ) 7.7 and 1.6 Hz, 1H), 7.46 (d, J ) 7.7 Hz, 1H), 7.35
(td, J ) 7.7 and 1.6 Hz, 1H), 2.30 (s, 3H); 13C NMR δ 192.4, 138.6,
135.4, 134.2, 133.7, 127.8, 124.8, 5.8; 77Se NMR δ 259.5 (lit. 26728).
Anal. Calcd for C8H8OSe: C, 48.26; H, 4.05. Found: C, 48.47; H,
4.11. A similar procedure was applied to 1e* for the synthesis of 17O-
labeled 1f (1f*). Spectral data for 1f*: 17O NMR δ 561.6 (broad s).
17O-Labeled Bis[2-(hydroxymethyl)phenyl] Diselenide (2e*). 17O-
labeled 1e (1e*) (56.2 mg, 0.15 mmol) was dissolved in MeOH (3
mL) under nitrogen. To the solution was added sodium borohydride
(174 mg, 4.6 mmol). The reaction mixture was stirred under nitrogen
for 2h and then under air overnight. The reaction was worked up with
aqueous sodium hydrogencarbonate, and the mixture was extracted with
CH2Cl2. The organic layer was collected, dried over anhydrous sodium
sulfate, filtrated, and then evaporated. Product 2e* (54.9 mg, 97%) was
obtained as yellow crystals in pure form, which was confirmed by
comparison of the NMR spectra with those of 2e.8a Spectral data for
2e: 1H NMR δ 7.68 (d, J ) 7.5 Hz, 2H), 7.39 (d, J ) 7.5 Hz, 2H),
7.31 (t, J ) 7.5 Hz, 2H), 7.20 (t, J ) 7.5 Hz, 2H), 4.73 (s, 4H); 13C
NMR δ 142.1, 135.0, 130.6, 128.9, 128.8, 128.4, 65.3; 77Se NMR δ
433.2. Anal. Calcd for C14H14O2Se2: C, 45.18; H, 3.79. Found: C,
44.88; H, 3.72. Spectral data for 2e*: 17O NMR δ 13.3 (broad s).
2-(Hydroxymethyl)benzeneselenenyl Chloride (2a). Compound 2a
was prepared from diselenide 2e quantitatively following the synthetic
procedure of 1a. Brown oil. Spectral data for 2a: 1H NMR δ 7.78
(broad m, 1H), 7.34 (broad m, 1H), 7.21 (broad m, 2H), 4.84 (broad s,
2H), 2.95 (broad s, 1H); 13C NMR δ 134.3, 130.0, 129.0, 127.1, 127.0,
125.9, 66.9; 77Se NMR δ 987.1. Spectral data for 17O-labeled 2a
(2a*): 17O NMR δ 24.9 (broad s).
2-(Hydroxymethyl)benzeneselenenyl Bromide (2b). Compound 2b
was prepared from diselenide 2e quantitatively following the synthetic
procedure of 1b. Brown oil. Spectral data for 2b: 1H NMR δ 7.85
(broad d, J ) 7.0 Hz, 1H), 7.34-7.22 (broad m, 3H), 4.85 (broad s,
2H), 2.39 (broad s, 1H); 13C NMR, signals not observed due to
significant line broadening; 77Se NMR δ 839.5. Spectral data for 17O-
labeled 2b (2b*): 17O NMR δ 23.2 (broad s).
2-(Hydroxymethyl)benzeneselenenyl Cyanate (2c). Compound 2c
was prepared from diselenide 2e in 72% yield following the synthetic
procedure of 1c. Colorless oil. Spectral data for 2c: 1H NMR δ 7.85-
7.82 (m, 1H), 7.32-7.27 (m, 2H), 7.23-7.20 (m, 1H), 4.71 (s, 2H);
13C NMR δ 139.5, 131.7, 129.4, 128.0, 127.8, 124.8, 104.7, 65.3; 77Se
NMR δ 314.5. Anal. Calcd for C8H7NOSe: C, 45.30; H, 3.33; N, 6.60.
Found: C, 45.26; H, 3.49; N 6.41. Spectral data for 17O-labeled 2c
(2c*): 17O NMR δ 22.0 (broad s).
worked up with aqueous sodium hydrogencarbonate, and the mixture
was extracted with CH2Cl2. Product 2f was obtained as a colorless oil
(25.6 mg, 96%). Spectral data for 2f: 1H NMR δ 7.43-7.40 (m, 1H),
7.36-7.33 (m, 1H), 7.24-7.20 (m, 2H), 4.72 (s, 2H), 2.40 (s, 1H),
2.31 (s, 3H); 13C NMR δ 140.8, 131.5, 130.6, 128.4, 128.0, 126.5, 7.3;
77Se NMR δ 157.2. Anal. Calcd for C8H10OSe: C, 47.77; H, 5.01.
Found: C, 47.71; H, 5.08. A similar procedure was applied to 2e* for
the synthesis of 17O-labeled 2f (2f*). Spectral data for 2f*: 17O NMR
δ 10.6 (broad s).
2-(Isopropoxymethyl)phenyl Isopropyl Selenide (4). To a dimethyl
sulfoxide (32 mL) solution of potassium hydroxide (3.60 g, 64 mmol)
were added compound 2e (743 mg, 2.0 mmol) and isopropyl bromide
(6.0 mL, 64 mmol) under nitrogen. After 20 h, isopropyl bromide (6.0
mL, 64 mmol) was added again to the solution. The mixture was stirred
for another 20 h. The reaction was worked up with water, and the
mixture was extracted with ether. Product 4 was obtained as a colorless
oil (508 mg, 47%) after purification of the crude product by silica gel
column chromatography (CH2Cl2). Spectral data for 4: 1H NMR δ 7.55
(dd, J ) 7.9 and 1.1 Hz, 1H), 7.46 (dd, J ) 7.9 and 1.1 Hz, 1H), 7.27
(td, J ) 7.9 and 1.1 Hz, 1H), 7.18 (td, J ) 7.9 and 1.1 Hz, 1H), 4.63
(s, 2H), 3.72 (sep, J ) 6.2 Hz, 1H), 3.47 (sep, J ) 6.2 Hz, 1H), 1.41
(d, J ) 6.2 Hz, 6H), 1.23 (d, J ) 6.2 Hz, 6H); 13C NMR δ 141.4,
134.7, 130.4, 128.5, 127.8, 127.4, 71.2, 70.3, 33.6, 24.1, 22.2; 77Se
NMR δ 368.5. Anal. Calcd for C13H20OSe: C, 57.56; H, 7.43. Found:
C, 57.64; H, 7.39. A similar procedure was applied to 2e* for the
synthesis of 17O-labeled 4 (4*). Spectral data for 4*: 17O NMR δ 39.0
(broad s).
Bis[2-(isopropoxymethyl)phenyl] Diselenide (3e). Compound 4
(465 mg, 1.7 mmol) was dissolved in MeOH. To the solution was added
30% hydrogen peroxide (1.5 mL, 15 mmol). After being stirred under
air for 17 h, the reaction mixture was evaporated in vacuo. The organic
residue was dissolved in dry THF (35 mL), and to the solution was
added lithium aluminum hydride (672 mg, 18 mmol). After having
refluxed for 2 h, the reaction was worked up with 2 M hydrochloric
acid (50 mL), and the mixture was extracted with ether (50 mL). Product
3e was obtained as a yellow oil (229 mg, 59%) after purification of
the crude product by silica gel column chromatography (CH2Cl2).
Spectral data for 3e: 1H NMR δ 7.75 (dd, J ) 7.9 and 1.1 Hz, 2H),
7.30 (dd, J ) 7.9 and 1.1 Hz, 2H), 7.20 (td, J ) 7.9 and 1.1 Hz, 2H),
7.17 (td, J ) 7.9 and 1.1 Hz, 2H), 4.58 (s, 2H), 3.72 (sep, J ) 6.2 Hz,
2H), 1.23 (d, J ) 6.2 Hz, 12H); 13C NMR δ 138.7, 132.4, 131.8, 128.7,
128.3, 127.2, 71.4, 70.5, 22.1; 77Se NMR δ 412.0. Anal. Calcd for
C20H26O2Se2: C, 52.64; H, 5.74. Found: C, 52.53; H, 5.44. A similar
procedure was applied to 4* for the synthesis of 17O-labeled 3e (3e*).
Spectral data for 3e*: 17O NMR δ 39.0 (broad s).
2-(Isopropoxymethyl)benzeneselenenyl Chloride (3a). Compound
3a was prepared from diselenide 3e quantitatively following the
synthetic procedure of 1a. Brown oil. Spectral data for 3a: 1H NMR
δ 7.77 (d, J ) 7.7 Hz, 1H), 7.34 (td, J ) 7.7 and 2.3 Hz, 1H), 7.20-
7.16 (m, 2H), 4.61 (s, 2H), 3.81 (sep, J ) 6.2 Hz, 1H), 1.27 (d, J )
6.2 Hz, 6H); 13C NMR δ 136.0, 134.6, 129.0, 128.0, 126.4, 126.1, 73.1,
71.0, 21.9; 77Se NMR δ 986.5. Spectral data for 17O-labeled 3a (3a*):
17O NMR δ 54.8 (broad s).
2-(Hydroxymethyl)benzeneselenenyl Phenyl Sulfide (2d). Com-
pound 2b, prepared from diselenide 2e (31.1 mg, 0.08 mmol) and
bromine (4.3 µL, 0.08 mmol), was dissolved in CH2Cl2 (3.5 mL). To
the solution were added pyridine (2 drops) and benzenethiol (19.0 µL,
0.19 mmol). After 1 h, the reaction mixture was concentrated under
reduced pressure to afford a yellow oil, which contained 2d, 2e, and
diphenyl disulfide. The yield of 2d was 25% according to the integration
2-(Isopropoxymethyl)benzeneselenenyl Bromide (3b). Compound
3b was prepared from diselenide 3e quantitatively following the
synthetic procedure of 1b. Brown oil. Spectral data for 3b: 1H NMR
δ 7.82 (d, J ) 7.7 Hz, 1H), 7.30 (td, J ) 7.7 and 1.7 Hz, 1H), 7.20 (t,
J ) 7.7 Hz, 1H), 7.16 (dd, J ) 7.7 and 1.7 Hz, 1H), 4.60 (s, 2H), 3.80
(sep, J ) 6.2 Hz, 1H), 1.26 (d, J ) 6.2 Hz, 6H); 13C NMR δ 136.9,
132.1, 131.3, 129.1, 126.8, 126.7, 72.6, 71.0, 22.0; 77Se NMR δ 857.8.
Spectral data for 17O-labeled 3b (3b*): 17O NMR δ 55.2 (broad s).
2-(Isopropoxymethyl)benzeneselenenyl Cyanate (3c). Compound
3c was prepared from diselenide 3e in 97% yield following the synthetic
procedure of 1c. Colorless oil. Spectral data for 3c: 1H NMR δ 7.83
(dd, J ) 7.5 and 1.3 Hz, 1H), 7.34-7.25 (m, 3H), 4.51 (s, 2H), 3.71
(sep, J ) 6.2 Hz, 1H), 1.23 (d, J ) 6.2 Hz, 6H); 13C NMR δ 137.7,
1
of H NMR. Due to the disproportionation of 2d, purification could
not be performed. Spectral data for 2d: 1H NMR δ 7.73 (d, J ) 8.5
Hz, 1H), 7.46 (d, J ) 8.5 Hz, 2H), 7.35 (d, J ) 8.5 Hz, 1H), 7.23-
7.18 (m, 5H), 4.77 (s, 2H); 13C NMR δ 140.4, 136.2, 132.4, 130.2,
129.5, 128.8, 128.4, 127.5, 127.2, 127.1, 64.5; 77Se NMR δ 501.8. A
similar procedure was applied to 2e* for the synthesis of 17O-labeled
2d (2d*). Spectral data for 2d*: 17O NMR δ 16.0 (broad s).
2-(Hydroxymethyl)phenyl Methyl Selenide (2f). Compound 2e
(24.7 mg, 0.07 mmol) and methyl iodide (20 µL, 0.32 mmol) were
dissolved in MeOH (2 mL) under nitrogen. To the solution was added
sodium borohydride (170 mg, 4.5 mmol). After 2 h, the reaction was
9
5316 J. AM. CHEM. SOC. VOL. 126, NO. 16, 2004

Nature of Nonbonded Se‚‚‚O Interactions
A R T I C L E S
131.6, 129.7, 128.8, 128.0, 125.4, 104.6, 71.8, 70.0, 21.8; 77Se NMR
δ 315.1. Anal. Calcd for C11H13NOSe: C, 51.98; H, 5.15; N, 5.51.
Found: C, 51.94; H, 5.07; N, 5.58. Spectral data for 17O-labeled 3c
(3c*): 17O NMR δ 44.6 (broad s).
(s, 2H), 3.73 (sep, J ) 6.2 Hz, 1H), 2.30 (s, 3H), 1.24 (d, J ) 6.2 Hz,
6H); 13C NMR δ 138.8, 132.8, 129.7, 128.5, 128.2, 125.9, 71.2, 70.0,
22.1, 6.2; 77Se NMR δ 166.6. Anal. Calcd for C11H16OSe: C, 54.32;
H, 6.63. Found: C, 54.45; H, 6.70. Spectral data for 17O-labeled 3f
(3f*): 17O NMR δ 39.0 (broad s).
2-(Isopropoxymethyl)benzeneselenenyl Isopropyl Sulfide (3d).
Compound 3d was prepared from diselenide 3e following the synthetic
procedure of 2d. Product 3d was obtained as a mixture with 3e and
diphenyl disulfide due to the disproportionation. The yield of 3d was
Computational Methods. All theoretical calculations were carried
out by using the Gaussian 98 program29 except for the atoms in
molecules (AIM) analysis. The hybrid Becke 3-Lee-Yang-Parr
(B3LYP) exchange-correlation functional30 was applied for DFT
calculations. Huzinaga’s 43321/4321/311 basis sets31 were used for Se
and Br, and 6-31G(d,p) basis sets were used for other atoms. The
combination is denoted here as 631H basis sets. Geometries were fully
optimized at the HF/631H and then B3LYP/631H levels of theory. To
save computational time, a phenylthio (SPh) group of compound d and
an arylseleno (SeAr) group of compound e were simplified to a
methylthio (SMe) group (in compound d′) and a methylseleno (SeMe)
group (in compound e′), respectively, in calculation. Similarly, an
isopropyl group of series 3 was simplified to a methyl group in series
5. For all stable conformers, the nature as a potential energy minimum
was established at the B3LYP/631H level by verifying that all
vibrational frequencies were real. The single-point energies were
corrected with zero-point energies (ZPE). The orbital interaction
energies between Se and O atoms (ESe‚‚‚O) as well as the atomic charges
(qSe and qO) were calculated by using the natural bond orbital (NBO)
method4 at the B3LYP/631H level. The AIM analysis was performed
by using the AIM2000 program.32 Topological properties of the electron
density (FSe‚‚‚O) at the bond critical points (BCP) of nonbonded Se‚‚‚O
interactions were characterized at the B3LYP/631H level.
1
48% according to the integration of H NMR. Spectral data for 3d:
1H NMR δ 7.83-7.15 (m, 9H), 4.56 (s, 2H), 3.73 (sep, J ) 6.2 Hz,
1H), 1.24 (d, J ) 6.2 Hz, 6H); 13C NMR δ 138.0, 134.7, 133.8, 129.9,
129.1, 128.9, 128.8, 128.4, 127.0, 126.8, 71.4, 70.4, 22.0; 77Se NMR
δ 503.8. Spectral data for 17O-labeled 3d (3d*): 17O NMR δ 44.4 (broad
s).
2-(Isopropoxymethyl)phenyl Methyl Selenide (3f). Compound 3f
was prepared from diselenide 3e in 52% yield following the synthetic
procedure of 1f. Colorless oil. Spectral data for 3f: 1H NMR δ 7.39
(dd, J ) 7.2 and 1.5 Hz, 1H), 7.36 (dd, J ) 7.2 and 1.5 Hz, 1H), 7.22
(td, J ) 7.2 and 1.5 Hz, 1H), 7.19 (td, J ) 7.2 and 1.5 Hz, 1H), 4.56
(29) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M.
A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A.; Stratmann,
R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin,
K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi,
R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.;
Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.;
Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz,
J. V.; Baboul, A. G.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.;
Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham,
M. A.; Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.;
Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.;
Gonzalez, C.; Head-Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian
98; Gaussian, Inc.: Pittsburgh, PA, 1998.
Acknowledgment. This work was supported by Grants in
Aid for Scientific Research Nos. 11120210 and 10133207 from
the Ministry of Education, Science, and Culture, Japan.
(30) (a) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785-789. (b)
Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5652.
(31) Gaussian Basis Sets for Molecular Calculations; Huzinaga, S., Ed.;
Elsevier: Amsterdam, 1984.
(32) Biegler-Ko¨nig, F.; Scho¨nbohm, J.; Bayles, D. J. Comput. Chem. 2001, 22,
545-559.
JA049690N
9
J. AM. CHEM. SOC. VOL. 126, NO. 16, 2004 5317